DESCRIPTION: (Applicant's Abstract) This proposed study focuses on one of society's most profound and disturbing problems: the co-occurrence of drug involvement, aggression, and depression among youth who are at-risk for dropping out of high school. This prevention trial examines the efficacy of a promising school-based, intervention program that responds to calls for comprehensive and multifocal interventions. The intervention model derives from etiologic models predictive of co-occurring adolescent drug involvement, depression, aggression, and school deviance and builds on multifaceted, theory-driven, school-based preventive interventions for high-risk youth. The PAYS (Parents and Youth with Schools) program combines the efficacious school-based Personal Growth Class with a novel parent component which includes home visit and parent group interventions to foster parenting skills and support strategies directed at re-establishing the youth-parent-school bond. All components integrate social network support and life skills training approaches to reduce risk factors (personal, family, school strain and deviant peer bonding) and enhance protective factors (life skills and prosocial peer, family and school bonding). The central hypothesis is that high-risk youth participating in the Project PAYS, compared to controls, will gain significantly in personal and social resources (mediating protective factors), and, thereby, demonstrate significantly greater reductions in drug involvement, aggression, depression and school deviance. A 2-group randomized experimental design is proposed with 7 repeated measurements over 25 months (2 pre-intervention, 2 during, 1 at-exit and 2 post-intervention) in 12 high schools. The sample size will be 744 high-risk youth (558 w/attrition). The design permits sophisticated assessment of group outcomes and trends over time. Latent growth models and structural equation modeling (with multiple indicators) will be used to examine variance/covariance structures and changes in outcomes and to test the posited mediating intervention effects. Furthermore, the proposed analytic strategies will allow exploration of key prevention research questions including the definition and outcome evaluation of youth with discrete risk vulnerabilities (including risk factor interactions) and the differential effects of gender and ethnicity. The research team is well-situated to carry out this research as it has 1) implemented and refined a successful indicated intervention program; 2) established procedures for identifying, recruiting, and retaining high-risk youth; and 3) generated and validated sensitive, developmentally appropriate instruments for measuring concepts central to the proposed intervention model. The study has both theory-testing and prevention research implications. It should markedly increase our understanding of the efficacy, feasibility, and acceptability of such multifocal, comprehensive interventions, and promote efforts to stem the progression of drug use and concomitant negative outcomes for high-risk youth.